GISdevelopment.net ---> AARS ---> ACRS 1991 ---> Education/Research

Appropriate education and training in Remote Sensing for the Asia/Pacific region

G.J.E. Hill, and S.R. Phinn
Department of Geographical Sciences
The University of Queensland
St Lucia Qld 4072

G.D. Kelly
Department of Lands
PO Box 40
Woolloongabba Qld 4102


Abstract
The disciplines of Remote Sensing has experienced rapid growth over the past decade. For the Asia/Pacific region, utilization of the technology ranges from heavy involvement in digital image processing and new generation sensor systems to very limited usage of any form of Remote Sensing, even aerial photography. Education and training programmes, therefore, need to be sensitive to the current status of Remote Sensing in individual countries. In particular emphasis should be given to applications where Remote Sensing is capable of making an important contribution to solving current problems. There is little point in educating personnel from the region in the use of inappropriate technology for applications of marginal interest.

Introduction
The roots of the discipline of Remote Sensing may be traced back to the invention of the camera in the late 1800s. Through the American Civil War and First World war the practical application of this new technology for earth reconnaissance was established, albeit for military purposes, and this eventually led to a dramatic increase in the use of Remote Sensing for civilian purposes (e.g. Colwell 1983). Aerial photography ushered in a new era of earth resource mapping and monitoring. With the advent of sophisticated scanner systems and satellite based platforms the term Remote Sensing became fashionable in the 1970s. While the tools of trade continue to develop, with new imaging systems and methods of analysis, the scope and purpose of applied Remote Sensing remains relatively unchanged. However, to-day's users of remotely sensed data are faced with a range of options that appear to increase

Knowledge/training required to participate in the technology as a whole has expanded greatly (e.g. Parry 1986).

These factors pose special problems for developing countries where the potential benefits of Remote Sensing are high but implementation is often difficult (e.g. Drury, 1990)

Under circumstances such as these, where resources are limited and information needs are high, it becomes critical for informed choices to be made regarding the implementation of Remote Sensing technology. Without a logical framework there is little likelihood that Remote Sensing will become an operational tool under the control of in-country, applied scientists. Decision of vital importance relate to :
  1. choice of suitable imagery;
  2. selection of projects; and
  3. training of personnel.
Making the right decisions regarding each of these factors should be guided by the current status of, and prospects for, Remote Sensing in the country involved. It will also involve a realistic assessment of the usefulness of the various forms of remotely sensed data in the local environment.

The aim of the current paper is to expand on the three elements listed above by drawing on the experience of the authors in implementing Remote Sensing technology in a range of countries from the south pacific region. Particular emphasis will be given to some of the difficulties involved in using satellite imagery in this part of the world and role of education and training in overcoming these.

Choice of suitable imagery
Matching imagery to projects would appear, at first glance, to be a straight forward task. The major variables of concern that relate to imagery would in include the following (e.g. Colwell 1983, Thaman 1986) Woodcock and Strahler 1987, Hill and Kelly 1989):
  1. spatial resolution and areal coverage
  2. spectral and radiometric characteristics
  3. availability
  4. cost
  5. facilities / expertise
  6. project type
Achieving a good match across this list provides ample opportunity for conflicts as far as satellite imagery is concerned. For example, for many of the South Pacific nations little scope exists for using satellite imagery for terrestrial applications. Land areas are so small, and land use/ land cover patterns heterogeneous that even imagery from the high resolution SPOT and Landsat TM systems is of limited usefulness (e.g. Parry 1986, Thaman 1986). Here, aerial photography is the most practical remotely sensed data to use.

To some extent, however there is an expectation on the part of the in-country user communities that the new generation, satellite technology must be best route to follow (for most application's areas). Remote Sensing specialists with career paths in scanner / digital processing areas may also have difficulty changing their ways and recommending less sophisticated forms of Remote Sensing. While "research projects" based on this philosophy may play a useful role in training personnel it is unlikely they will promote operational Remote Sensing.

Resolution is a factor that requires close attention in framing any Remote Sensing project in the south pacific. This, of course, does not mean that satellite data are not useful and appropriate for many tasks. However, the standard satellite Remote Sensing applications in continental areas are often not appropriate for island nations.

Lingering on aerial photography, it is clear that lack of recent coverage and limited availability are major drawbacks. However, for the south pacific, the repetitive coverage of the satellites is of academic interest only if real-time data are required. Cloud cover is a constant problem and programming requests running over years may be necessary if cloud free imagery is required for specific areas. Both the Asian and South Pacific regions experience greater that 50% cloudness during at least one season of the year (Parry 1986). Again this does not preclude the use of the data but it does restrict the range of applications that can be accomplished on the routine basis.

Selection of projects
Following from the above, it is clear that if satellite imagery is to be used careful attention must be paid to selection of projects. A perusal of the literature reporting use of satellite Remote Sensing in the South Pacific demonstrates clearly that the greatest scope exists for the marine and coastal environments (e.g Loubersac and Populus 1986, Bour 1988, Vercelli et al. 1988, Collate 1989).

Projects need to assess not only a matching of imagery to task, but also, the appropriateness of the methodology followed to the facilities and expertise present in the host environment. The Polynesian Remote Sensing facility, operated by IFREMER (French Oceanographic Institute) is located in Pepeete, Tahiti and another Remote Sensing laboratory is operated in Noumea, New Caledonia by ORSTROM (French Scientific Research Institute for Co-operative Development). Most of the digital processing work emanating from the region originates from these two facilities. Digital processing facilities also exist in Fiji where there is a growing involvement in Remote Sensing activities. For most of the region, however, image processing activities are completed "off-shore". Given the realities of economic and practical circumstances, it is probable that most countries in the regional will not have direct access to image processing facilities for some time.

To address these limitations a workshop conducted by the Economic and Social Commission for development in Asia and the Pacific (ESCAP), United Nations Development Programme (UNDP), Regional Remote Sensing Programme recommended.

    "…the establishment of a Pacific regional Remote Sensing and geographic information systems facility to foster and encourage the co-ordination and development of this technology within the region." (ESCAP 1989:12).
Specific resource requirements for this suggested facility were identified and further development of the concept supported in the "PACNET" proposal for a regional Remote Sensing and GIS network for the Pacific (Mau 1991). This proposal has defined organizational structure and establishment criteria and highlighted priority needs in training and funding.

Training of personnel
One unfortunate reality of technology transfer within the Asia/Pacific region is that wastage occurs through lack of facilities within individual organizations and countries. This is particularly true for the upper end of the education spectrum. Highly motivated and competent individuals complete advanced training in state of the art image processing but return to work environments where no suitable facilities exist. In some cases there is little chance that such facilities will ever be in place. At the opposite end of the spectrum there may be a marked lack of facilities and training for the use of relatively unsophisticated, though very practical, data such as aerial photography and photographic imagery (as opposed to digital products).

Such wastage is damaging both to the persons involved, their countries, and the profile of Remote Sensing. There is scope here for regional educational institutions to modify course structures to help rectify this situation for south pacific students. The university of the south pacific will also beed to develop programmes in Remote Sensing if the discipline is to eventually become an integral part of resource assessment and management activities in the region.

These problems have been addressed by organizations such as ESCAP (ESCAP 1986) and its regional Remote Sensing programme (ESCAP 1990). There are models available in the region of structured training and technology implementation programmes and selection of projects that are geared too existing facilities and levels of expertise (e.g. see Kelly et al. 1991). However, this is by no means uniform across the region.

Discussion
The foregoing provides a guide to some of the difficulties currently being faced in providing appropriate Remote Sensing technology for the south pacific region.

As far as the task of promoting the advantages Remote Sensing has to offer is concerned, it is crucial to appreciate that the physical, economic and social environment of the area is unique. Programmes for the implementation of Remote Sensing must therefore be guided by these factors and not by current practice for other parts of the globe.

This would appear particularly relevant to strategic planning for education and training. The Remote Sensing products recommended (choice of imagery) and the formulation of suitable demonstration and operational projects (selection of projects will differ substantially from the packages that form the basis of typical training suites offered elsewhere. This also implies that training in the use of technology that is not available and not likely to be available in the foreseeable future is of little benefit. Activities must be designed to reflect these differences and the current, as well as projected status of Remote Sensing in the various countries.

There would appear to be large challenges facing education/ training programme in Remote Sensing for the region. Not because of a lack of activity per see but because of a lack of facilities to support the knowledge obtained by those gaining skills in Remote Sensing. Unless these programmes are linked to a parallel development of facilities and services in Remote Sensing must of the impetus given through education will be lost.

References
  • Bour, W. (1988). SPOT images for coral reef mapping in New Caledonia. A fruitful approach for classic and new topics. Proc. 6th Internat. Coral Reef Symposium. Townsville,
  • Collate, P. (1989) The use of satellite imagery for coastal and nearshore applications in the south pacific. In : ESCAP/UNDP (1989) Remote Sensing for land and sea resources. Report of the workshop on Remote Sensing for land and sea resource surveys and Evaluation in the pacific: Applications, Co-ordination and Training. Port Villa, Vanuatu, pp 49-57.
  • Colwell, R.N. (1983) Manual of Remote Sensing (2nd Ed.) Volume 1 American Society of Photogrmmetry, falls Church, Virginia.
  • Drury, S.A (1990) Manual of Remote Sensing : Interpreting images of the Earth. Oxford Scientific Publications, New York.
  • ESCAP (1986) Report on the Expert Group meeting o Remote Sensing training and education in the ESCAP region, Bangkok, Thailand.
  • ESCAP/UNDP (1989) Remote Sensing for Land and sea Resources. Report of the workshop on Remote Sensing for land and sea resource surveys and evaluation in the Pacific: Applications, co-ordination and Training. Port Villa, Vanuatu, 7 - 11 August.
  • ESCAP (1990) Regional Remote Sensing Programme, South Pacific Sub-programme: Project findings and recommendations. ESCAP/UNDP Regional Remote Sensing programme, Bangkok, Thailand.
  • ESCAP/UNDP (1991) Remote Sensing and geographic information systems for the pacific. Report of the working group meeting on regional co-operation in applications for Remote Sensing and geographic information systems technology for the south Pacific. Suva, fiji, 18-22 March.
  • Hill, G.J.E an Kelly, G.D. (1989) Strategic Planning for the implementation of operational Remote Sensing : Training, education and project support. in : ESCAP/UNDP (1989) Remote Sensing for land and sea resource surveys and evaluation in the pacific: Applications, Co-ordination and training. Port Villa, Vanuatu, pp 73-79.
  • Kelly, G.D. Tuiloma, P. and Hill, G.J.E 91991) The Remote Sensing Training component of the North West Viti Levu Pilot project Fiji. Asian Pacific Remote Sensing Journal.
  • Loubersac, L. and Populus, J.91986). The Application of high resolution satellite data for coastal management and planning in a pacific coral island. Geocarto, 2:17-31
  • Mau, G (1991) PACNET: proposal for a Pacific Remote Sensing and GIS Network. In : ESCAP/UNDP (1991) Remote Sensing and geographic information systems for the pacific. Report of the working group meeting on regional co-operation in applications for Remote and geographic information systems technology for the South Pacific. Suva, Fiji, 18-22 March, pp 78-90.
  • Parry, J.T. (1986) Background, perspective and issues for Remote Sensing in the tropics. In : Eden, M.J. and Parry, J.T. (Eds) Remote Sensing and Tropical Land Management. J. Wiley and Sons, London, UK, pp 337-360.
  • Thaman, R.R. (1986) Coming down to earth ; strengthening Remote Sensing applications in Island countries. In :ESCAP/UNDP 91986) Remote Sensing in the Pacific Islands. Combined reports of the Pacific Island regional Remote Sensing workshop and training course on resource mapping. Suva, Fiji, pp 49-50.
  • Vercelli, C., Gabrie, C and Ricard, M. (1988). Utilization of SPOT-1 data in coral reef cartography, Moorea Island and Takapoto Atoll, French Polynesia. Proc. 6th Internat. Coral Reef Symposium. Townsville, Autralia., pp. 463-68.
  • Woodcock, C.E and Starhler, A.H (1988) The factor of scale in Remote Sensing. Remote Sensing of Environment, 21 : 311-332.